Slope monitoring and early warning system based on single-mode multi-terminal positioning
By adopting single-mode multi-terminal positioning technology and isolated forest algorithm in the slope monitoring and early warning system, combined with Beidou satellite positioning, the problems of high complexity and low fault tolerance of the existing system are solved, and landslide monitoring and early warning is realized when the base station signal is unavailable , improves the accuracy and reliability of the monitoring system.
Patent Information
- Application Number
- CN202410838377.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-26
- Publication Date
- 2024-11-08
AI Technical Summary
The existing slope monitoring and early warning system has high complexity and low fault tolerance, which leads to the inability to continue to perform displacement monitoring and early warning when the base station is damaged, and cannot effectively prevent secondary landslides.
Adopt a slope monitoring and early warning system based on single-mode and multi-terminal positioning, and use RTK monitoring module, acquisition module, processing module, evaluation module and early warning module, combined with the isolated forest algorithm and Beidou satellite positioning system to realize differential positioning and abnormal displacement evaluation to ensure that Landslide monitoring and early warning can still be carried out normally when the base station signal is unavailable.
It improves the accuracy of slope displacement information and the reliability of monitoring systems, significantly improves the warning level of geological risks and the safety performance of equipment, and reduces potential risks.
Smart Images

Figure CN118918689A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of slope monitoring, and in particular to a slope monitoring and early warning system based on single-mode multi-terminal positioning. Background Art
[0002] In recent years, due to the reduction of shallow and easily exploitable resources, large-scale open-pit mines have gradually transitioned to deep open-pit mining, which has become the main development trend of open-pit mining worldwide. However, as the height and angle of the slopes of open-pit mines increase, the risk of landslides and safety accidents also increases significantly. In order to reduce the threat of these risks to safety and property, most mines have introduced a complete set of displacement monitoring and alarm systems. The current displacement monitoring and early warning process includes sensor data acquisition, data transmission through transmission interfaces, controllers, and repeaters to the base station, and then transmitted to cloud servers and other terminals for data analysis and alarm warning. However, this process has the problems of high complexity and low fault tolerance, and once a landslide occurs, the base station may be damaged, which makes the equipment unable to continue to perform displacement monitoring and early warning, and cannot effectively prevent secondary landslides. Summary of the invention
[0003] In order to overcome the above-mentioned deficiencies of the prior art, the present invention provides a slope monitoring and early warning system based on single-mode multi-terminal positioning.
[0004] The technical solution adopted by the present invention is: a slope monitoring and early warning system based on single-mode multi-terminal positioning, including multiple monitoring poles and 5G base stations arranged on the target slope, each of the poles is provided with an RTK monitoring module, an acquisition module, a processing module, an evaluation module, an early warning module, and a transmission module, the RTK monitoring module is connected to the processing module and the transmission module, the acquisition module is used to collect the carrier phase signal of the monitoring pole other than the monitoring pole where it is located, and transmit the collected information to the processing module, the processing module calculates the displacement of the monitoring pole where it is located according to the received carrier phase signal, and realizes differential positioning according to the carrier phase signal transmitted by the base station or the monitoring pole other than the monitoring pole where it is located, the evaluation module determines the abnormal displacement value according to the calculation result of the processing module and the displacement information of the monitoring pole other than the monitoring pole where it is located collected by the acquisition module, and transmits the information to the early warning module, the early warning module makes an early warning process according to the information it receives, and the transmission module transmits the displacement information of the monitoring pole where it is located to the monitoring pole other than the monitoring pole where it is located.
[0005] Furthermore, the RTK monitoring module includes two receivers, a transmitting station and a receiving station. One of the two receivers serves as a base station and the other serves as a mobile station. When there is no signal, the base station transmits the signal to the monitoring pole other than the monitoring pole where it is located through the transmission module, and serves as the base station of the monitoring pole other than the monitoring pole where it is located. The mobile station receives the radio information of the 5G base station and the satellite signal transmitted by the Beidou positioning system through the receiving station.
[0006] Furthermore, the evaluation module adopts an isolation forest algorithm to construct an isolated tree to evaluate the displacement information of the monitoring pole where it is located and determine the abnormal displacement value.
[0007] Furthermore, the evaluation module performs the evaluation including the following steps:
[0008] Step 1: Calculate the current sensor position information based on the Beidou satellite carrier phase difference principle:
[0009] {(x1,y1,z1),(x2,y2,z2),…,(x n ,y n , z n )}
[0010] And the sensor position information at the previous moment:
[0011] {(x1′, y1′, z1′), (x2′, y2′, z2′),…, (x n ′,y n ′,z n ′)}
[0012] Where n is the number of monitoring poles.
[0013] Step 2: Calculate the displacement information of each monitoring rod: (X i ,Y i ,Z i )
[0014] Where X i =x i ′-x i , Y i =y i ′-y i , Z i =z i ′-z i , (i=1,2,…,n), integrated into data set D
[0015] D={(x1,y1,z1),…,(x n-1 ,y n-1 , z n-1 ),…,(x n ,yn , z n )}
[0016] Step 3: Data collation, Z-Score standardization of data:
[0017] X i ′=(X i -E(X)) / Var(X),
[0018] Y i ′=(Y i -E(Y)) / Var(Y),
[0019] Z i ′=(Z i -E(Z)) / Var(Z)
[0020] At this time, the i-th data point of the data set D′ is (X i ,Y i ,Z i ).
[0021] Step 4: Build multiple isolated trees, each of which is a binary tree. The tree construction process is the same as that of the two-dimensional feature space.
[0022] Step 5: Calculate the path length. For the i-th data point, calculate its path length in the isolated tree, that is, the number of edges required to reach the data point from the root node of the tree, expressed as H i =h(X i ′,Y i ′,Z i ′).
[0023] Step 6: Calculate the isolation index, traverse each isolation tree, and calculate the depth of data point i in each tree. The isolation index ISO_i can be expressed as:
[0024]
[0025] Where c(n) = E(h(X i ′,Y i ′,Z i ′))
[0026] Step 7: Calculate the anomaly score by normalizing the isolation index to obtain the anomaly score 1 / ISO_i.
[0027] Step 8: Set the threshold, the actual output range of the anomaly score is [-0.5, 0.5]
[0028] Step 9: Make a judgment. According to the actual output range of the anomaly score in the previous step, make the following judgment:
[0029] 1) If the returned anomaly score is between [-0.5, -0.2], the point is an abnormal sample;
[0030] 2) If the returned anomaly score is between [0.2, 0.5], it is considered a normal sample;
[0031] 3) If the anomaly scores returned by all samples are between (-0.2, 0.2), it means that there are no obvious outliers in the entire data set and they are considered as outliers.
[0032] Furthermore, in step 4 of the evaluation module performing the evaluation, the step of constructing the tree is:
[0033] a. Randomly select a feature: Randomly select a feature from the three features X′, Y′, and Z′, such as X′;
[0034] b. Randomly select a split value: Randomly select a value between the minimum and maximum values of X as the split threshold.
[0035] c. Split the data points into two subsets: According to the selected feature segmentation threshold X′, the data points (X′, Y′, Z′) are divided into two subsets, one subset contains data points less than the threshold, and the other subset contains data points greater than the threshold.
[0036] d. Recursively build the tree: Repeat the above steps, recursively dividing the subsets until the tree reaches the maximum depth.
[0037] The present invention adopts an isolated tree algorithm in the evaluation module to construct an isolated tree to evaluate the displacement information of the equipment, determine the abnormal displacement value, and make the slope displacement information more accurate; by setting two receivers in the RTK monitoring module, one of the receivers acts as a reference station, and transmits the signal to other devices through the transmission module when there is no signal, acting as a reference station for other devices. The mobile station receives the radio information of the 5G base station and the satellite signal transmitted by the Beidou positioning system through the receiving radio station, so that when a landslide causes the base station to be damaged and unable to transmit, the system can still perform landslide monitoring and early warning normally, and can effectively warn of secondary landslides, etc. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] Figure 1 It is a schematic diagram of the structure of the present invention.
[0039] Figure 2 This is a schematic diagram of the isolated forest tree structure of the present invention.
[0040] Figure 3 Schematic diagram of the evaluation module steps of the present invention. DETAILED DESCRIPTION
[0041] In order to more clearly understand the technical solution of the present invention, the present invention is further described below in conjunction with the accompanying drawings.
[0042] like Figure 1-Figure 3 As shown, a slope monitoring and early warning system based on single-mode multi-terminal positioning includes multiple monitoring poles and 5G base stations arranged on the target slope, each pole is provided with an RTK monitoring module, an acquisition module, a processing module, an evaluation module, an early warning module, and a transmission module, the RTK monitoring module is connected with the processing module and the transmission module, the acquisition module is used to collect the carrier phase signals of the monitoring poles other than the monitoring pole where it is located, and transmit the collected information to the processing module, the processing module calculates the displacement of the monitoring pole where it is located according to the received carrier phase signals, and uses the advanced Signal processing technologies such as waveform analysis, digital filtering and fast Fourier transform are used to extract and analyze the carrier phase information in the satellite signal, and differential positioning is achieved according to the carrier phase information transmitted by the base station or other monitoring pole transmission modules, thereby further improving the accuracy of position measurement. The evaluation module determines the abnormal displacement value and transmits the information to the early warning module according to the calculation results of the processing module and the displacement information of the monitoring poles other than the monitoring pole where the evaluation module is located collected by the acquisition module. The early warning module makes an early warning processing according to the information it receives, and the transmission module transmits the displacement information of the monitoring pole where the evaluation module is located to the monitoring poles other than the monitoring pole where the evaluation module is located.
[0043] Specifically, the RTK monitoring module includes two receivers, a transmitting station and a receiving station. One of the two receivers serves as a base station and the other serves as a mobile station. When there is no signal, the base station transmits the signal to the monitoring poles other than the monitoring pole where it is located through the transmission module, serving as the base station for the monitoring poles other than the monitoring pole where it is located. The mobile station receives the radio information of the 5G base station and the satellite signals transmitted by the Beidou positioning system through the receiving station.
[0044] Preferably, the evaluation module adopts an isolation forest algorithm to construct an isolated tree to evaluate the displacement information of the monitoring pole where the isolated tree is located, and determine the abnormal displacement value.
[0045] Specifically, the evaluation module performs evaluation including the following steps:
[0046] Step 1: Calculate the current sensor position information based on the Beidou satellite carrier phase difference principle:
[0047] {(x1,y1,z1),(x2,y2,z2),…,(x n ,y n , z n )}
[0048] And the sensor position information at the previous moment:
[0049] {(x1′, y1′, z1′), (x2′, y2′, z2′),…, (x n ′,y n ′,z n ′)}
[0050] Where n is the number of monitoring poles.
[0051] Step 2: Calculate the displacement information of each monitoring rod: (X i ,Y i ,Z i )
[0052] Where X i =x i ′-x i , Y i =y i ′-y i , Z i =z i ′-z i , (i=1,2,…,n), integrated into data set D
[0053] D={(x1,y1,z1),…,(x n-1 ,y n-1 , z n-1 ),…,(x n ,y n , z n )}
[0054] Step 3: Data collation, Z-Score standardization of data:
[0055] X i ′=(X i -E(X)) / Var(X),
[0056] Y i ′=(Y i -E(y)) / Var(y),
[0057] Z i ′=(Zi-E(z)) / Var(z)
[0058] At this time, the i-th data point of the data set D is (X i ,Y i ,Z i ).
[0059] Step 4: Build multiple isolated trees, each of which is a binary tree. The tree construction process is the same as that of the two-dimensional feature space.
[0060] Step 5: Calculate the path length. For the i-th data point, calculate its path length in the isolated tree, that is, the number of edges required to reach the data point from the root node of the tree, expressed as H i =h(X i ′,Y i ′,Z i ′).
[0061] Step 6: Calculate the isolation index, traverse each isolation tree, and calculate the depth of data point i in each tree. The isolation index ISO_i can be expressed as:
[0062]
[0063] Where c(n) = E(h(X i ′,Y i ′,Z i ′))
[0064] Step 7: Calculate the anomaly score by normalizing the isolation index to obtain the anomaly score 1 / ISO_i.
[0065] Step 8: Set the threshold. The actual output range of the anomaly score is [-0.5, 0.5]. By looking at the underlying code and official documentation, it is found that the original calculation result score_samples of the outlier value of the isolation forest algorithm in the scikit-learn machine learning library is exactly the opposite of the result of the anomaly score in the iForest paper, and the value range is [-1, 0]. In the called decision function, there is an algorithm offset when calculating the anomaly score, which defaults to -0.5. Therefore, the actual numerical output range of the isolation forest anomaly score in this paper is [-0.5, 0.5].
[0066] Step 9: Make a judgment based on the actual output range of the anomaly score in the previous step.
[0067] 1) If the returned anomaly score is between [-0.5, -0.2], the point is an abnormal sample;
[0068] 2) If the returned anomaly score is between [0.2, 0.5], it is considered a normal sample;
[0069] 3) If the anomaly scores returned by all samples are between (-0.2, 0.2), it means that there are no obvious outliers in the entire data set and they are considered as outliers.
[0070] Specifically, in step 4 of the evaluation module performing the evaluation, the step of constructing the tree is:
[0071] a. Randomly select a feature: Randomly select a feature from the three features X′, Y′, and Z′;
[0072] b. Randomly select a split value: Randomly select a value between the minimum and maximum values of X as the split threshold.
[0073] c. Split the data points into two subsets: According to the selected feature segmentation threshold X′, the data points (X′, Y′, Z′) are divided into two subsets, one subset contains data points less than the threshold, and the other subset contains data points greater than the threshold.
[0074] d. Recursively build the tree: Repeat the above steps, recursively dividing the subsets until the tree reaches the maximum depth.
[0075] The present invention can accurately predict when 5G base station signals are available and unavailable, thereby significantly improving the early warning level of geological risks and the safety performance of equipment. By comprehensively applying these technologies and functions, the efficiency and reliability of slope monitoring can be effectively improved, and potential risks can be reduced.
[0076] The above description is only a preferred embodiment of the present invention, so all equivalent changes or modifications made according to the structure, characteristics and principles described in the scope of the patent application of the present invention are included in the scope of the patent application of the present invention.
Claims
1. A slope monitoring and early warning system based on single-mode multi-terminal positioning, comprising a plurality of monitoring poles and a 5G base station arranged on a target slope, characterized in that: Each of the poles is provided with an RTK monitoring module, an acquisition module, a processing module, an evaluation module, an early warning module, and a transmission module. The RTK monitoring module is connected to the processing module and the transmission module. The acquisition module is used to collect carrier phase signals of monitoring poles other than the monitoring pole where it is located, and transmit the collected information to the processing module. The processing module calculates the displacement of the monitoring pole where it is located according to the received carrier phase signal, and realizes differential positioning according to the carrier phase signal transmitted by the base station or the monitoring pole other than the monitoring pole where it is located. The evaluation module determines the abnormal displacement value according to the calculation result of the processing module and the displacement information of the monitoring poles other than the monitoring pole where it is located collected by the acquisition module, and transmits the information to the early warning module. The early warning module makes early warning processing according to the information it receives, and the transmission module transmits the displacement information of the monitoring pole where it is located to the monitoring poles other than the monitoring pole where it is located.
2. A slope monitoring and early warning system based on single-mode multi-terminal positioning, characterized in that: The RTK monitoring module includes two receivers, a transmitting radio station and a receiving radio station. One of the two receivers serves as a base station and the other serves as a mobile station. When there is no signal, the base station transmits the signal to the monitoring poles other than the monitoring pole where it is located through the transmission module, and serves as the base station of the monitoring poles other than the monitoring pole where it is located. The mobile station receives the radio information of the 5G base station and the satellite signal transmitted by the Beidou positioning system through the receiving radio station.
3. The slope monitoring and early warning system based on single-mode multi-terminal positioning according to claim 1 is characterized in that: The evaluation module adopts an isolation forest algorithm to construct an isolated tree to evaluate the displacement information of the monitoring pole where it is located and determine the abnormal displacement value.
4. The slope monitoring and early warning system based on single-mode multi-terminal positioning according to claim 3 is characterized in that: The evaluation module includes the following steps: Step 1: Calculate the current sensor position information based on the Beidou satellite carrier phase difference principle: {(x1,y1,z1),(x2,y2,z2),…,(x n y n ,z n )} And the sensor position information at the previous moment: {(x1′,y1′,z1′),(x2′,y2′,z2′),…,(x n ′,y n ′,z n ′)} Where n is the number of monitoring poles; Step 2: Calculate the displacement information of each monitoring rod: (X i ,Y i ,Z i ) Where X i =x i ′-x i , Y i =y i ′-y i , Z i =z i ′-z i , (i=1,2,…,n), integrated into data set D, D.{(x1,y1,z1),...,(x n-1 y n-1 ,z n-1 ),...,(x n y n ,z n )}? Step 3: Data collation, Z-Score standardization of data: X i ′=(X i -E(X)) / Var(X), AND i ′=(And i -E(Y)) / Var(Y), With i ′=(Z i -E(Z)) / Var(Z), At this time, the i-th data point of the data set D′ is (X i ′,Y i ′,Z i ′); Step 4: Build multiple isolated trees, each of which is a binary tree. The tree construction process is the same as that of the two-dimensional feature space. Step 5: Calculate the path length. For the i-th data point, calculate its path length in the isolated tree, that is, the number of edges required to reach the data point from the root node of the tree, expressed as H i =h(X i ′,Y i ′,Z i ′). Step 6: Calculate the isolation index, traverse each isolation tree, and calculate the depth of data point i in each tree. The isolation index ISO_i can be expressed as: Where c(n) = E(h(X i ′,Y i ′,Z i ′)) Step 7: Calculate the anomaly score by normalizing the isolation index to obtain the anomaly score 1 / ISO_i. Step 8: Set the threshold, the actual output range of the anomaly score is [-0.5, 0.5] Step 9: Make a judgment based on the actual output range of the anomaly score in the previous step. 1) If the returned anomaly score is between [-0.5, -0.2], the point is an abnormal sample; 2) If the returned anomaly score is between [0.2, 0.5], it is considered a normal sample; 3) If the anomaly scores returned by all samples are between (-0.2, 0.2), it means that there are no obvious outliers in the entire data set and they are considered as outliers.
5. The slope monitoring and early warning system based on single-mode multi-terminal positioning according to claim 4 is characterized in that: In step 4 of the evaluation performed by the evaluation module, the steps of constructing a tree are: a. Randomly select a feature: Randomly select a feature from the three features X′, Y′, and Z′; b. Randomly select a split value: Randomly select a value between the minimum and maximum values of the selected feature as the split threshold. c. Split the data points into two subsets: According to the selected feature segmentation threshold, the data points (X′, Y′, Z′) are divided into two subsets, one subset contains data points less than the threshold, and the other subset contains data points greater than the threshold. d. Recursively build the tree: Repeat the above steps, recursively dividing the subsets until the tree reaches the maximum depth.
